Parabolic Experimentation with AEO: A Deep Dive into Ramp's Growth Strategy

Parabolic Experimentation with AEO: A Deep Dive into Ramp's Growth Strategy

This presentation by George Bonaci, VP of Growth at Ramp, titled "Parabolic Experimentation with AEO (Answer Engine Optimization)," details Ramp's journey and experimental approach to achieving visibility in the age of generative AI.

The core premise is that the goal of marketing has shifted from traditional SEO ranking to ensuring that when a user asks AI who to trust, the answer is your brand [00:59].

The Shift to the "New Trust Interface"

Bonaci argues that AI answers are the new trust interface [02:06]. He highlights that users, even those unfamiliar with the technology (like his retired father), use AI heavily because the friction is so low. While people know AI can "hallucinate," the ease of getting an answer makes them use it regardless, prioritizing convenience over perfection [02:34]. This necessitates a focus on AEO (Answer Engine Optimization) over traditional SEO.

Ramp's Experimental Journey to AEO

Ramp's AEO strategy was driven by a series of high-velocity, rigorous experiments, often in partnership with Profound:

1. Targeting Low Search Volume Keywords

  • Hypothesis: Can content deemed "worthless" by traditional SEO tools be valuable for LLMs?

  • Action: Revamped programmatic content focused on very low monthly search volume keywords [04:01].

  • Result: The LLMs rewarded the effort with huge increases in citations and visibility [04:32], confirming a fundamental difference between AEO and SEO.

2. Long-Form Plugs and "Jobs to be Done"

  • Action: Extended content depth for low-volume keywords, focusing on a specific human pain point or a "job to be done" (JTD) [05:07]. Data was integrated into the narrative.

  • Result: Visibility for the test content went up by about 25% [05:44].

3. The Proprietary Data Unlock (Ramp Rate)

After an initial experiment failed by going too narrow and losing sight of the human element [05:56], Ramp realized that LLMs heavily favored unique, proprietary data [06:39].

Ramp leveraged its unique access to transaction data from over 40,000 customers to create a product called Ramp Rate [07:01].

  • Ramp Rate's Purpose: A product built primarily for LLM consumption to systematically output insights derived from vendor spend data (e.g., the top 10 most expensive software or price trends) [08:02].

  • Ramp Rate's Anatomy: Each page is structured to serve both machine and human: a time-series chart, a narrative summary (with an in-house economist's editorial), and a raw data table for ingestion by LLMs [09:01].

  • Ramp Rate's Results: This approach led to a roughly 150% increase in session growth to these pages [07:30], proving that fresh, proprietary, and crawlable data wins [07:51].

Three Major Takeaways for AEO Strategy

Based on their experimental journey, Ramp established three key strategies for the audience:

  1. Fresh Data Really Wins: You must invest in systems to keep information updated on a weekly or monthly basis. Historical data does not provide the same spike in relevance and citations as constantly refreshed information [13:42].

  2. Name Your Metric: Define a proprietary, opinionated metric that is unique to your company and inherently citable by an LLM (e.g., Ramp Rate). This is the fundamental question to answer before starting any AEO project [13:59].

  3. Design a Loop: Build automated systems and products that create a positive feedback loop. Scalable solutions, like Ramp Rate, create a vast surface area for continuous experimentation, which in turn drives more citations and increased visibility [14:32].

AEO, Growth, ProfoundFrancesca Tabor